Hugging Face's logo

Heather-Driver
/
distilbert-classn-LinearAlg-finetuned-pred-span-width-5

Transformers
TensorBoard
Safetensors
distilbert
Generated from Trainer
Model card Files
xet
Metrics Community
Heather-Driver commited on
Commit
5bce81c
·
verified ·
1 Parent(s): cbb87c3

Classification Training

Browse files
Files changed (1) hide show
  1. README.md +41 -41
README.md CHANGED
@@ -21,11 +21,11 @@ should probably proofread and complete it, then remove this comment. -->
21
 
22
  This model is a fine-tuned version of [dslim/distilbert-NER](https://huggingface.co/dslim/distilbert-NER) on the None dataset.
23
  It achieves the following results on the evaluation set:
24
- - Loss: 0.7740
25
- - Accuracy: 0.8333
26
- - F1: 0.8294
27
- - Precision: 0.8449
28
- - Recall: 0.8333
29
 
30
  ## Model description
31
 
@@ -60,42 +60,42 @@ The following hyperparameters were used during training:
60
 
61
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
62
  |:-------------:|:-------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
63
- | 4.9685 | 0.6849 | 50 | 2.4456 | 0.0714 | 0.0561 | 0.1349 | 0.0714 |
64
- | 4.8271 | 1.3699 | 100 | 2.4172 | 0.0714 | 0.0666 | 0.1461 | 0.0714 |
65
- | 4.8954 | 2.0548 | 150 | 2.3559 | 0.0794 | 0.0889 | 0.1505 | 0.0794 |
66
- | 4.7314 | 2.7397 | 200 | 2.2811 | 0.1587 | 0.1514 | 0.2272 | 0.1587 |
67
- | 4.5724 | 3.4247 | 250 | 2.1496 | 0.2857 | 0.2585 | 0.3108 | 0.2857 |
68
- | 4.2953 | 4.1096 | 300 | 1.9594 | 0.5 | 0.4699 | 0.5189 | 0.5 |
69
- | 3.7997 | 4.7945 | 350 | 1.7529 | 0.5873 | 0.5732 | 0.5979 | 0.5873 |
70
- | 3.3537 | 5.4795 | 400 | 1.5588 | 0.6349 | 0.6191 | 0.6491 | 0.6349 |
71
- | 2.8409 | 6.1644 | 450 | 1.3903 | 0.6667 | 0.6559 | 0.7118 | 0.6667 |
72
- | 2.4597 | 6.8493 | 500 | 1.1847 | 0.7143 | 0.7040 | 0.7239 | 0.7143 |
73
- | 1.9307 | 7.5342 | 550 | 1.0572 | 0.7460 | 0.7435 | 0.7809 | 0.7460 |
74
- | 1.4402 | 8.2192 | 600 | 0.9328 | 0.7619 | 0.7580 | 0.7878 | 0.7619 |
75
- | 1.1287 | 8.9041 | 650 | 0.8659 | 0.7619 | 0.7557 | 0.7991 | 0.7619 |
76
- | 0.831 | 9.5890 | 700 | 0.8162 | 0.7937 | 0.7899 | 0.8000 | 0.7937 |
77
- | 0.575 | 10.2740 | 750 | 0.7284 | 0.8333 | 0.8331 | 0.8469 | 0.8333 |
78
- | 0.3999 | 10.9589 | 800 | 0.7241 | 0.8254 | 0.8230 | 0.8352 | 0.8254 |
79
- | 0.2961 | 11.6438 | 850 | 0.6876 | 0.8095 | 0.8054 | 0.8189 | 0.8095 |
80
- | 0.1947 | 12.3288 | 900 | 0.6900 | 0.8333 | 0.8293 | 0.8453 | 0.8333 |
81
- | 0.1269 | 13.0137 | 950 | 0.7272 | 0.8254 | 0.8204 | 0.8379 | 0.8254 |
82
- | 0.102 | 13.6986 | 1000 | 0.7102 | 0.8254 | 0.8234 | 0.8398 | 0.8254 |
83
- | 0.0692 | 14.3836 | 1050 | 0.7489 | 0.8333 | 0.8296 | 0.8439 | 0.8333 |
84
- | 0.0753 | 15.0685 | 1100 | 0.7253 | 0.8333 | 0.8295 | 0.8437 | 0.8333 |
85
- | 0.043 | 15.7534 | 1150 | 0.7408 | 0.8254 | 0.8213 | 0.8302 | 0.8254 |
86
- | 0.0256 | 16.4384 | 1200 | 0.7352 | 0.8333 | 0.8305 | 0.8444 | 0.8333 |
87
- | 0.0242 | 17.1233 | 1250 | 0.7523 | 0.8333 | 0.8293 | 0.8453 | 0.8333 |
88
- | 0.0167 | 17.8082 | 1300 | 0.7792 | 0.8413 | 0.8370 | 0.8548 | 0.8413 |
89
- | 0.0108 | 18.4932 | 1350 | 0.7845 | 0.8175 | 0.8114 | 0.8315 | 0.8175 |
90
- | 0.0205 | 19.1781 | 1400 | 0.7618 | 0.8333 | 0.8294 | 0.8449 | 0.8333 |
91
- | 0.0095 | 19.8630 | 1450 | 0.7952 | 0.8254 | 0.8209 | 0.8427 | 0.8254 |
92
- | 0.0073 | 20.5479 | 1500 | 0.7598 | 0.8254 | 0.8215 | 0.8309 | 0.8254 |
93
- | 0.0161 | 21.2329 | 1550 | 0.7812 | 0.8254 | 0.8206 | 0.8374 | 0.8254 |
94
- | 0.0073 | 21.9178 | 1600 | 0.7782 | 0.8333 | 0.8293 | 0.8453 | 0.8333 |
95
- | 0.0063 | 22.6027 | 1650 | 0.7674 | 0.8333 | 0.8294 | 0.8449 | 0.8333 |
96
- | 0.0068 | 23.2877 | 1700 | 0.7631 | 0.8333 | 0.8294 | 0.8449 | 0.8333 |
97
- | 0.0053 | 23.9726 | 1750 | 0.7727 | 0.8333 | 0.8294 | 0.8449 | 0.8333 |
98
- | 0.0055 | 24.6575 | 1800 | 0.7740 | 0.8333 | 0.8294 | 0.8449 | 0.8333 |
99
 
100
 
101
  ### Framework versions
 
21
 
22
  This model is a fine-tuned version of [dslim/distilbert-NER](https://huggingface.co/dslim/distilbert-NER) on the None dataset.
23
  It achieves the following results on the evaluation set:
24
+ - Loss: 0.6839
25
+ - Accuracy: 0.8254
26
+ - F1: 0.8236
27
+ - Precision: 0.8363
28
+ - Recall: 0.8254
29
 
30
  ## Model description
31
 
 
60
 
61
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
62
  |:-------------:|:-------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
63
+ | 4.9451 | 0.6849 | 50 | 2.4266 | 0.1190 | 0.0734 | 0.0812 | 0.1190 |
64
+ | 4.9004 | 1.3699 | 100 | 2.4006 | 0.1508 | 0.1049 | 0.1791 | 0.1508 |
65
+ | 4.864 | 2.0548 | 150 | 2.3589 | 0.1667 | 0.1427 | 0.1829 | 0.1667 |
66
+ | 4.7694 | 2.7397 | 200 | 2.3013 | 0.1984 | 0.1656 | 0.1611 | 0.1984 |
67
+ | 4.5881 | 3.4247 | 250 | 2.2257 | 0.2778 | 0.2715 | 0.3458 | 0.2778 |
68
+ | 4.4949 | 4.1096 | 300 | 2.0636 | 0.4127 | 0.3981 | 0.4152 | 0.4127 |
69
+ | 4.0569 | 4.7945 | 350 | 1.8632 | 0.5397 | 0.5396 | 0.5936 | 0.5397 |
70
+ | 3.6327 | 5.4795 | 400 | 1.6784 | 0.6190 | 0.6196 | 0.6836 | 0.6190 |
71
+ | 3.0577 | 6.1644 | 450 | 1.4586 | 0.6429 | 0.6206 | 0.6410 | 0.6429 |
72
+ | 2.6585 | 6.8493 | 500 | 1.2315 | 0.7063 | 0.7024 | 0.7198 | 0.7063 |
73
+ | 2.0628 | 7.5342 | 550 | 1.0891 | 0.7381 | 0.7300 | 0.7656 | 0.7381 |
74
+ | 1.5864 | 8.2192 | 600 | 0.9558 | 0.7857 | 0.7781 | 0.8529 | 0.7857 |
75
+ | 1.1035 | 8.9041 | 650 | 0.8837 | 0.7698 | 0.7657 | 0.8141 | 0.7698 |
76
+ | 0.8962 | 9.5890 | 700 | 0.8059 | 0.8254 | 0.8178 | 0.8573 | 0.8254 |
77
+ | 0.6185 | 10.2740 | 750 | 0.7363 | 0.8492 | 0.8527 | 0.8948 | 0.8492 |
78
+ | 0.4703 | 10.9589 | 800 | 0.6929 | 0.8254 | 0.8237 | 0.8539 | 0.8254 |
79
+ | 0.3438 | 11.6438 | 850 | 0.6574 | 0.8175 | 0.8192 | 0.8409 | 0.8175 |
80
+ | 0.2744 | 12.3288 | 900 | 0.6597 | 0.8175 | 0.8131 | 0.8335 | 0.8175 |
81
+ | 0.1704 | 13.0137 | 950 | 0.6842 | 0.8175 | 0.8188 | 0.8592 | 0.8175 |
82
+ | 0.1469 | 13.6986 | 1000 | 0.6285 | 0.8333 | 0.8286 | 0.8475 | 0.8333 |
83
+ | 0.0849 | 14.3836 | 1050 | 0.6737 | 0.8095 | 0.8112 | 0.8460 | 0.8095 |
84
+ | 0.1058 | 15.0685 | 1100 | 0.6356 | 0.8413 | 0.8383 | 0.8545 | 0.8413 |
85
+ | 0.069 | 15.7534 | 1150 | 0.6495 | 0.8333 | 0.8364 | 0.8672 | 0.8333 |
86
+ | 0.0304 | 16.4384 | 1200 | 0.6442 | 0.8492 | 0.8484 | 0.8687 | 0.8492 |
87
+ | 0.0548 | 17.1233 | 1250 | 0.6309 | 0.8413 | 0.8366 | 0.8560 | 0.8413 |
88
+ | 0.0388 | 17.8082 | 1300 | 0.6645 | 0.8254 | 0.8258 | 0.8468 | 0.8254 |
89
+ | 0.0108 | 18.4932 | 1350 | 0.6785 | 0.8413 | 0.8380 | 0.8581 | 0.8413 |
90
+ | 0.0396 | 19.1781 | 1400 | 0.6720 | 0.8175 | 0.8196 | 0.8410 | 0.8175 |
91
+ | 0.0237 | 19.8630 | 1450 | 0.6676 | 0.8333 | 0.8328 | 0.8440 | 0.8333 |
92
+ | 0.0084 | 20.5479 | 1500 | 0.6876 | 0.8254 | 0.8237 | 0.8389 | 0.8254 |
93
+ | 0.0451 | 21.2329 | 1550 | 0.6760 | 0.8333 | 0.8333 | 0.8497 | 0.8333 |
94
+ | 0.0204 | 21.9178 | 1600 | 0.6818 | 0.8333 | 0.8333 | 0.8497 | 0.8333 |
95
+ | 0.0151 | 22.6027 | 1650 | 0.6830 | 0.8095 | 0.8099 | 0.8276 | 0.8095 |
96
+ | 0.02 | 23.2877 | 1700 | 0.6841 | 0.8254 | 0.8237 | 0.8389 | 0.8254 |
97
+ | 0.0057 | 23.9726 | 1750 | 0.6829 | 0.8254 | 0.8236 | 0.8363 | 0.8254 |
98
+ | 0.006 | 24.6575 | 1800 | 0.6839 | 0.8254 | 0.8236 | 0.8363 | 0.8254 |
99
 
100
 
101
  ### Framework versions